Instructions to use predator3769/peft-dialogue-summary-training-1767960897 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use predator3769/peft-dialogue-summary-training-1767960897 with PEFT:

from peft import PeftModel
from transformers import AutoModelForCausalLM

base_model = AutoModelForCausalLM.from_pretrained("microsoft/phi-2")
model = PeftModel.from_pretrained(base_model, "predator3769/peft-dialogue-summary-training-1767960897")

Transformers

How to use predator3769/peft-dialogue-summary-training-1767960897 with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="predator3769/peft-dialogue-summary-training-1767960897")

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("predator3769/peft-dialogue-summary-training-1767960897", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use predator3769/peft-dialogue-summary-training-1767960897 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "predator3769/peft-dialogue-summary-training-1767960897"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "predator3769/peft-dialogue-summary-training-1767960897",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/predator3769/peft-dialogue-summary-training-1767960897

SGLang

How to use predator3769/peft-dialogue-summary-training-1767960897 with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "predator3769/peft-dialogue-summary-training-1767960897" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "predator3769/peft-dialogue-summary-training-1767960897",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "predator3769/peft-dialogue-summary-training-1767960897" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "predator3769/peft-dialogue-summary-training-1767960897",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use predator3769/peft-dialogue-summary-training-1767960897 with Docker Model Runner:
```
docker model run hf.co/predator3769/peft-dialogue-summary-training-1767960897
```

peft-dialogue-summary-training-1767960897

This model is a fine-tuned version of microsoft/phi-2 on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.2741

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 0.0002
train_batch_size: 1
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 4
optimizer: Use OptimizerNames.PAGED_ADAMW_8BIT with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 1
training_steps: 1000

Training results

Training Loss	Epoch	Step	Validation Loss
1.8756	0.0665	25	0.8069
0.5969	0.1331	50	0.6492
0.7373	0.1996	75	0.5183
0.4293	0.2661	100	0.5101
0.611	0.3327	125	0.4577
0.3935	0.3992	150	0.4664
0.5535	0.4657	175	0.4308
0.3765	0.5323	200	0.4349
0.487	0.5988	225	0.4017
0.3464	0.6653	250	0.4049
0.4799	0.7319	275	0.3849
0.3313	0.7984	300	0.3953
0.4697	0.8649	325	0.3686
0.354	0.9315	350	0.3699
0.3705	0.9980	375	0.3521
0.3968	1.0639	400	0.3441
0.2887	1.1304	425	0.3566
0.3884	1.1969	450	0.3338
0.2879	1.2635	475	0.3366
0.3961	1.3300	500	0.3237
0.3062	1.3965	525	0.3233
0.3572	1.4631	550	0.3145
0.2787	1.5296	575	0.3168
0.3594	1.5961	600	0.3096
0.2788	1.6627	625	0.3116
0.3406	1.7292	650	0.3015
0.2848	1.7957	675	0.3008
0.3488	1.8623	700	0.2946
0.2774	1.9288	725	0.2960
0.3092	1.9953	750	0.2888
0.3158	2.0612	775	0.2855
0.2678	2.1277	800	0.2856
0.2927	2.1943	825	0.2854
0.276	2.2608	850	0.2819
0.3042	2.3273	875	0.2790
0.2351	2.3939	900	0.2770
0.305	2.4604	925	0.2759
0.2599	2.5269	950	0.2753
0.3014	2.5935	975	0.2747
0.2236	2.6600	1000	0.2741

Framework versions

PEFT 0.18.0
Transformers 4.57.3
Pytorch 2.8.0+cu126
Datasets 4.4.2
Tokenizers 0.22.1

Downloads last month: 9

Model tree for predator3769/peft-dialogue-summary-training-1767960897

Base model

microsoft/phi-2

Adapter

(992)

this model